David L. Braff

Pharmacological studies of prepulse inhibition models of sensorimotor gating deficits in schizophrenia: a decade in review.

Abstract.Rationale: Patients with schizophrenia exhibit deficits in an operational measure of sensorimotor gating: prepulse inhibition (PPI) of startle. Similar deficits in PPI are produced in rats by pharmacological or developmental manipulations. These experimentally induced PPI deficits in rats are clearly not animal models of schizophrenia per se, but appear to provide models of sensorimotor gating deficits in schizophrenia patients that have face, predictive, and construct validity. In rodents, disruptions in PPI of startle are produced by: stimulation of D2 dopamine (DA) receptors, produced by amphetamine or apomorphine; activation of serotonergic systems, produced by serotonin (5-HT) releasers or direct agonists at multiple serotonin receptors; and blockade of N-methyl-D-aspartate (NMDA) receptors, produced by drugs such as phencyclidine (PCP). Accordingly, dopaminergic, serotonergic, and glutamatergic models of disrupted PPI have evolved and have been applied to the identification of potential antipsychotic treatments. In addition, some developmental manipulations, such as isolation rearing, have provided non-pharmacological animal models of the PPI deficits seen in schizophrenia. Objective: This review summarizes and evaluates studies assessing the effects of systemic drug administrations on PPI in rats. Methods: Studies examining systemic drug effects on PPI in rats prior to January 15, 2001 were compiled and organized into six annotated appendices. Based on this catalog of studies, the specific advantages and disadvantages of each of the four main PPI models used in the study of antipsychotic drugs were critically evaluated. Results: Despite some notable inconsistencies, the literature provides strong support for significant disruptions in PPI in rats produced by DA agonists, 5-HT2 agonists, NMDA antagonists, and isolation rearing. Each of these models exhibits sensitivity to at least some antipsychotic medications. While the PPI model based on the effects of direct DA agonists is the most well-validated for the identification of known antipsychotics, the isolation rearing model also appears to be sensitive to both typical and atypical antipsychotics. The 5-HT PPI model is less generally sensitive to antipsychotic medications, but can provide insight into the contribution of serotonergic systems to the actions of newer antipsychotics that act upon multiple receptors. The deficits in PPI produced by NMDA antagonists appear to be more sensitive to clozapine-like atypical antipsychotics than to typical antipsychotics. Hence, despite some exceptions to this generalization, the NMDA PPI model might aid in the identification of novel or atypical antipsychotic medications. Conclusions: Studies of drug effects on PPI in rats have generated four distinctive models that have utility in the identification of antipsychotic medications. Because each of these models has specific advantages and disadvantages, the choice of model to be used depends upon the question being addressed. This review should help to guide such decisions.

Dopaminergic stimulation disrupts sensorimotor gating in the rat.

Prepulse inhibition is a cross-species phenomenon in which reflex responses to discrete sensory events are modified by weak prestimulation. In experiments designed to investigate the neuropharmacological mechanism of this form of information processing, and its relevance to schizophrenic psychopathology, apomorphine (0.125–4.0 mg/kg) and d-amphetamine (0.5–4.0 mg/kg) were administered to rats in an attempt to modify prepulse inhibition of the acoustic startle response. Rats were presented with 40 ms, 118 dB[A] acoustic pulses which were intermittently preceded by a weak 80 dB[A] acoustic prepulse. Both apomorphine and d-amphetamine induced a significant loss of prepulse inhibition, as reflected by increased pulse-preceded-by-prepulse versus pulse-alone startle magnitudes. Haloperidol (0.1 mg/kg), a specific D2 dopamine receptor antagonist, prevented the effects of 2.0 mg/kg apomorphine on prepulse inhibition, while having little effect by itself. An additional study investigated the effects of chronic intermittent administration of 2.5 mg/kg d-amphetamine. Rats given amphetamine for 8 consecutive days also displayed a loss of prepulse inhibition, with no evidence of tolerance. Finally, prepulse inhibition was examined under high- and low-intensity startle stimulus conditions; apomorphine (1.0 mg/kg) induced a loss of prepulse inhibition under both intensity conditions in approximately equal proportion. The results of these studies suggest a connection between sensorimotor gating, as measured by prepulse inhibition, and dopaminergic overactivity, supporting suggestions that information processing deficits in schizophrenia may be responsible for some psychotic symptoms and their effective treatment by antipsychotic D2 dopamine antagonists.

Is it possible to be schizophrenic yet neuropsychologically normal

This study identified and characterized a group of schizophrenic patients without neuropsychological (NP) impairment. A comprehensive NP battery was administered to 171 schizophrenic outpatients and 63 normal comparison participants. Each participants NP status was classified through blind clinical ratings by 2 experienced neuropsychologists; 27% of the schizophrenics were classified as NP normal. The NP-normal and NP-impaired schizophrenics were similar in terms of most demographic, psychiatric, and functional characteristics, except that NP-normal patients had less negative and extrapyramidal symptoms, were on less anticholinergic medication, socialized more frequently, and were less likely to have had a recent psychiatric hospitalization. The existence of NP-normal schizophrenics suggests that the pathophysiology underlying the cognitive deficits often associated with schizophrenia may be distinct from that causing some of its core psychiatric features.

Startle response models of sensorimotor gating and habituation deficits in schizophrenia

Studies of prepulse inhibition and habituation of startle responses elicited by intense stimuli provide some unusual opportunities for cross-species explorations of attentional deficits characteristic of schizophrenic patients. Schizophrenic patients exhibit deficits in both the prepulse inhibition of startle and the habituation of startle. The behavioral plasticity of startle responses and the comparability of the test paradigms used in rats and humans greatly facilitates the development of animal models of specifiable behavioral abnormalities in schizophrenic patients. This review describes two such examples of parallel animal and human models, one involving sensorimotor gating and the other examining behavioral habituation. Evidence is presented supporting the involvement of mesolimbic dopaminergic systems in the modulation of prepulse inhibition or sensorimotor gating and the importance of central serotonergic systems in the habituation of startle.

Neonatal excitotoxic hippocampal damage in rats causes post-pubertal changes in prepulse inhibition of startle and its disruption by apomorphine.

Neonatal excitotoxic hippocampal damage in the rat results in postpubertal onset of a variety of abnormal behaviors related to excessive dopaminergic transmission in the mesolimbic/nigrostriatal system, and thus may be considered an animal model of some aspects of schizophrenia. Because sensorimotor gating is impaired in adult patients with schizophrenia and in rats with experimentally induced mesolimbic dopamine hyperactivity, the present experiments investigated the effects of neonatal (postnatal day 7, PD7) ibotenic acid (3 µg) lesions of the ventral hippocampus (VH) on the amplitude and prepulse inhibition (PPI) of acoustic startle in prepubertal (PD35) and postpubertal (PD56) rats. Startle was elicited using 105 and 118-dB pulses alone or preceded by 4, 8, or 16 dB above-background prepulses in rats treated with vehicle or apomorphine (APO; 0.025 or 0.1 mg/kg SC). At PD35, PPI in VH-lesioned rats did not differ significantly from these measures in sham operated rats. Apomorphine significantly increased startle amplitude and reduced PPI in both sham operated and VH-lesioned rats at PD35. At PD56, startle amplitude in VH-lesioned rats was not significantly different from controls, but PPI was reduced significantly compared to controls. Ventral hippocampus lesioned rats also exhibited an exaggerated reduction in PPI after treatment with APO. These findings provide further evidence of postpubertal impairments that may be related to increased mesolimbic dopamine transmission and receptor sensitivity in rats with neonatal hippocampal damage, and provide further support for the fidelity of this animal model of schizophrenia.

Startle gating deficits occur across prepulse intensities in schizophrenic patients

The effects of prepulse stimuli of different intensities in inhibiting the startle reflex was assessed in 14 age-matched and gender-matched schizophrenic patients and 14 normal controls. The subjects were presented with startling stimuli consisting of bursts of white noise (106 dBA) with or without prepulse stimuli. Four intensities of prepulse stimuli were utilized: 75, 80, 85, and 90 dBA. Throughout the testing, the background noise was maintained at 70 dBA. The prepulse stimuli more effectively inhibited the startle reflex in the control group compared to the schizophrenic patients who showed deficient prepulse inhibition (gating) of the startle reflex. These results suggest that schizophrenics have impaired central inhibitory mechanisms over a fairly broad range of background noise to prepulse ratios. Further studies are needed to clarify exactly which ratios are optimal in eliciting prepulse inhibition (PPI) and in differentiating between schizophrenic and control groups.

Social cognition in schizophrenia: Relationships with neurocognition and negative symptoms

Despite the growing importance of social cognition in schizophrenia, fundamental issues concerning the nature of social cognition in schizophrenia remain unanswered. One issue concerns the strength of the relationships between social cognition and key features of the disorder such as neurocognitive deficits and negative symptoms. The current study employed structural equation modeling to examine three key questions regarding the nature of social cognition in schizophrenia: 1) Are social cognition and neurocognition in schizophrenia better modeled as one or two separate constructs? 2) Are social cognition and negative symptoms in schizophrenia better modeled as one or two separate constructs?, and 3) When social cognition, neurocognition, and negative symptoms are included in a single model, is social cognition more closely related to neurocognition or to negative symptoms? In this cross sectional study, one hundred outpatients with schizophrenia or schizoaffective disorder were administered measures of social cognition, neurocognition, and negative symptoms. A two-factor model that represented social cognition and neurocognition as separate constructs fit the data significantly better than a one-factor model, suggesting that social cognition and neurocognition are distinct, yet highly related, constructs. Likewise, a two-factor model that represented social cognition and negative symptoms as separate constructs fit the data significantly better than a one-factor model, suggesting that social cognition and negative symptoms are distinct constructs. A three-factor model revealed that the relationship between social cognition and neurocognition was stronger than the relationship between social cognition and negative symptoms. The current findings start to provide insights into the structure of social cognition, neurocognition, and negative symptoms in schizophrenia.

Gamma band oscillations reveal neural network cortical coherence dysfunction in schizophrenia patients.

BACKGROUND Gamma band activity has been associated with many sensory and cognitive functions, and is important for cortico-cortical transmission and the integration of information across neural networks. The aims of the present study were to determine if schizophrenia patients have deficits in the generation and maintenance of coherent, synchronized oscillations in response to steady-state stimulation, and to examine the clinical and cognitive correlates of the electroencephalography (EEG) oscillatory dynamics. METHODS Schizophrenia patients (n = 100) and nonpsychiatric subjects (n = 80) underwent auditory steady-state event-related potential testing. Click trains varying in rate of stimulation (20, 30, and 40 Hz) were presented; EEG-evoked power and intertrial phase synchronization were obtained in response to each stimulation frequency. Subjects also underwent clinical and neurocognitive assessments. RESULTS Patients had reductions in both evoked power and phase synchronization in response to 30- and 40-Hz stimulation but normal responsivity to 20-Hz stimulation. Maximal deficits were detected in response to 40-Hz stimulation. A modest association of reduced working memory performance and 40-Hz intertrial phase synchronization was present in schizophrenia patients (r = .32, p <.01). CONCLUSIONS Schizophrenia patients have frequency-specific deficits in the generation and maintenance of coherent gamma-range oscillations, reflecting a fundamental degradation of basic integrated neural network activity.

The nature of learning and memory impairments in schizophrenia.

The California Verbal Learning Test was used to characterize the learning and memory impairment in schizophrenia (SC) and to evaluate potential clinical and demographic factors associated with this impairment. SC patients (n = 175) performed worse than normal comparison (NC) subjects (n = 229) on all learning, recall, and recognition memory measures. The most important clinical correlates of these impairments were earlier age of onset, more negative symptoms, and greater anticholinergic medication dosage. SC patients showed a prominent retrieval deficit as indicated by disproportionate improvement when tested in a recognition, rather than a free recall, format. A residual impairment seen with recognition testing suggests a mild encoding deficit as well. In contrast, the relative absence of a storage deficit is suggested by the lack of rapid forgetting. Using a discriminant function analysis that differentiates cortical dementia [i.e., Alzheimers disease (AD)], subcortical dementia [i.e., Huntingtons disease (HD)], and normals, it was found that 50% of the SC patients were classified as having a subcortical memory profile and 35% were classified as having a normal profile, whereas only 15% were classified as having a cortical memory profile. Although these findings reflect the clinical heterogeneity often found in SC, results suggest that most SC patients demonstrate a pattern of learning and memory impairments that resembles the pattern seen in patients with primary subcortical (specifically striatal) pathology.

Schizophrenic-like sensorimotor gating abnormalities in rats following dopamine infusion into the nucleus accumbens

Previous studies have demonstrated that several dopamine agonists disrupt sensorimotor gating as measured by prepulse inhibition (PPI) of the acoustic startle response (ASR) in rats. Schizophrenic patients also exhibit deficits in PPI when the prepulse preceeds the startle stimulus by less than 500 ms. In experiment 1, dopamine (0–40 µg) infused directly into the nucleus accumbens in rats caused a dose-dependent decrease in PPI at prepulse intervals shorter than 500 ms. In experiment 2, this effect of accumbens dopamine infusions on sensorimotor gating was found to vary with changes in prepulse intensity. These findings strongly suggest that increased mesolimbic dopamine activity is one substrate of the sensorimotor gating deficits in rats that are caused by treatment with dopamine agonists; similar substrates might mediate deficits in PPI exhibited by schizophrenic patients.